A digital asynchronous logic is proposed as a generic matrix readout for Monolithic active pixel sensors. The architecture is implemented for pixel pitch ranging from 18 to 30 mu m. Post-layout simulations with realistic hit shapes and rates up to 200 MHz/cm2 2 show that time stamping at the 20 ns level can be achieved for a digital power cost below 10 mW/cm2. 2 .

In this paper, two of the most common calibration methods of synchronous TDCs, which are the bin-by-bin calibration and the average-bin-width calibration, are first presented and compared. Then, an innovative new robust calibration method for asynchronous TDCs is proposed and evaluated. Simulation results showed that: (i) For a synchronous TDC, the bin-by-bin calibration, applied to a histogram, does not improve the TDC’s differential non-linearity (DNL); nevertheless, it improves its Integral Non-Linearity (INL), whereas the average-bin-width calibration significantly improves both the DNL and the INL. (ii) For an asynchronous TDC, the DNL can be improved up to 10 times by applying the bin–by-bin calibration, whereas the proposed method is almost independent of the non-linearity of the TDC and can improve the DNL up to 100 times. The simulation results were confirmed by experiments carried out using real TDCs implemented on a Cyclone V SoC-FPGA. For an asynchronous TDC, the proposed calibration method is 10 times better than the bin-by-bin method in terms of the DNL improvement.

Time-resolved fluorescence measurement screening and sorting has been proven to be interesting for biological applications. Fluorescence lifetime (FLT) measurement provides additional information to the measurement of the intensity or polarization of the fluorescence emission. Furthermore, droplet microfluidics enables higher throughput than well plate readers. Few researches have been recently done about the microfluidic droplet sorting based on the fluorescence lifetime measurement. Some of the proposed solutions support high throughput but they do not extract the FLT directly from the fluorescence decay. In this paper, we present an alternative low-cost system for the microfluidic droplets sorting. We implemented a FLT measurement system based on the time-correlated single photon counting (TCSPC) technique on a cyclone V SoC-FPGA. For the excitation light source, we use a simple pulsed laser diode and a single photon avalanche diode (SPAD) as a photodetector. The optical system is an ad hoc microscope. The droplet generation is done with flow focusing technique in a PDMS-based microfluidics chip. This system was successfully tested in real-time at a droplet rate of more than 3000 droplets per second.

Field-programmable gate array (FPGA)-based time-to-digital converters (TDCs) suffer from large bin width variations. This issue imposes performing a calibration process to compensate the nonlinearity of the TDC. The most commonly used calibration technique is the bin–by-bin calibration that can improve the differential nonlinearity (DNL) of the TDC up to 10 times. In this paper, we propose a new robust calibration method for asynchronous TDCs and compare it with the bin-by-bin method. The simulation results showed that the proposed method is less sensitive to the nonlinearity of the TDC, and an improvement of more than 100 times can be achieved in regard to the DNL of the non-calibrated TDC. These results have been confirmed by experimental measurements made on an asynchronous TDC implemented on a Cyclone V SoC-FPGA kit. Density code tests were performed to measure the root mean square (RMS) DNL of the TDC and those of the calibrated histograms for the two methods. For a TDC with a DNL of 0.71 least significant bit (LSB), the DNL of the calibrated histogram is 0.053 LSB for the classical bin-by-bin method and 0.005 LSB, i.e., 10 times better, for the proposed method.

The usage of single-photon avalanche diode arrays is becoming increasingly common in various domains such as medical imaging, automotive vision systems, and optical communications. Nowadays, thanks to the development of microelectronics technologies, the SPAD arrays designed for these applications has been drastically well-facilitated, allowing for the manufacturing of large matrices. However, there are growing challenges for the design of readout circuits with the needs of reducing their energy consumption (linked to the usage cost) and data rate. Indeed, the design of the readout circuit for the SPAD array is generally based on synchronous logic; the latter requires synchronization that may increase the dead time of the SPADs and clock trees management that are known to increase power consumption. With these limitations, the long-neglected asynchronous (clockless) logic proved to be a better alternative because of its ability to operate without a clock. In this paper, we presented the design of a 16-to-1 fixed-priority tree arbiter readout circuit for a SPAD array based on asynchronous logic principles. The design of this circuit was explained in detail and supported by simulation results. The manufactured chip was tested, and the experimental results showed that it is possible to record up to 333 million events per second; no reading errors were detected during the data extraction test.

This work presents a Fluorescence Life-Time (FLT) measurement system for real-time microfluidic droplet sorting in high throughput conditions. This system is implemented using a low cost System-on-Chip (SoC) Field-Programmable Gate Array (FPGA) platform, that combines a Cyclone V FPGA with a dual-core ARM Cortex-a9 Hard Processor System (HPS). A time-correlated single photon counting system is implemented in the FPGA part and the data are transferred to the SDRAM of the HPS part to be processed by a developed bare-metal C program to extract the FLT of each droplet passing through the detection spot. According to the droplet's measured FLT, an action could be taken to sort this droplet. The system automatically detects the droplets and extracts their FLT values at different simulated droplet flow rates; from a few droplets up to 1 thousand droplets per second. Thanks to the use of a maximum Likelihood-based algorithm, the standard deviation of the measured FLTs of simulated droplets of the same material is only 30% above the theoretical quantum photon shot noise limit.

There are hundreds of research publications that theoretically discuss the implementation of Tapped Delay Line based Time to Digital Converters (TDL TDCs) on Field-Programmable Gate Array (FPGA) targets. However, most of these works do not cover the timing issues that will be encountered mostly due to the routing delays. The purpose of this work is to highlight the main timing issues that should be considered when implementing TDCs in FPGA targets and propose practical approaches to overcome these issues. As a study case, a full design methodology of a TDC on a Cyclone V FPGA target is presented in this work.

Novel imaging methods permitting real-time, wide-field and quantitative optical mapping of biological tissue properties offer an unprecedented range of potential new applications for clinical use such as guided surgery or patient monitoring. However, significant technical challenges have so far prevented such tools from performing in real-time (for both acquisition and processing) and therefore from being deployed in clinical practice. To overcome these limitations, recent research introduced methods based on Spatial Frequency Domain Imaging (SFDI) that allow real-time (within milliseconds) wide-field imaging of optical properties. In this study we present a novel implementation of general purpose graphic processing unit (GPGPU) direct programming in C CUDA (Compute Unified Device Architecture) for real-time, wide-field and quantitative multispectral imaging using our recently-developed spatio-temporal modulation of light imaging method. Using this new method, we are able to quantitatively obtain optical properties images (1 megapixel) at 2 wavelengths (665 nm and 860 nm) in only 1.52 ms with at most 1% error in comparison with standard Matlab processing.

Imaging methods permitting real-time, wide-field, and quantitative optical mapping of biological tissue properties offer an unprecedented range of applications for clinical use. Following the development of spatial frequency domain imaging, we introduce a real-time demodulation method called single snapshot of optical properties (SSOPs). However, since this method uses only a single image to generate absorption and reduced scattering maps, it was limited by a degraded image quality resulting in artifacts that diminished its potential for clinical use. We present filtering strategies for improving the image quality of optical properties maps obtained using SSOPs. We investigate the effect of anisotropic two-dimensional filtering strategies for spatial frequencies ranging from 0.1 to 0.4 mm(-1) directly onto N = 10 hands. Both accuracy and image quality of the optical properties are quantified in comparison with standard, multiple image acquisitions in the spatial frequency domain. Overall, using optimized filters, mean errors in predicting optical properties using SSOP remain under 8.8% in absorption and 7.5% in reduced scattering, while significantly improving image quality. Overall this work contributes to advance real-time, wide-field, and quantitative diffuse optical imaging toward clinical evaluation. (C) The Authors. Published by SPIE under a Creative Commons Attribution 4.0 Unported License.

We present the effects of using a single-pixel camera approach to extract optical properties with the single-snapshot spatial frequency-domain imaging method. We acquired images of a human hand for spatial frequencies ranging from 0.1 to 0.4  mm  -  1 with increasing compression ratios using adaptive basis scan wavelet prediction strategy. In summary, our findings indicate that the extracted optical properties remained usable up to 99% of compression rate at a spatial frequency of 0.2  mm  -  1 with errors of 5% in reduced scattering and 10% in absorption.

The development of real-time, wide-field and quantitative diffuse optical imaging methods is becoming increasingly popular for biological and medical applications. Recent developments introduced a novel approach for real-time multispectral acquisition in the spatial frequency domain using spatio-temporal modulation of light. Using this method, optical properties maps (absorption and reduced scattering) could be obtained for two wavelengths (665 nm and 860 nm). These maps, in turn, are used to deduce oxygen saturation levels in tissues. However, while the acquisition was performed in real-time, processing was performed post-acquisition and was not in real-time. In the present article, we present CPU and GPU processing implementations for this method with special emphasis on processing time. The obtained results show that the proposed custom direct method using a General Purpose Graphic Processing Unit (GPGPU) and C CUDA (Compute Unified Device Architecture) implementation enables 1.6 milliseconds processing time for a 1 Mega-pixel image with a maximum average error of 0.1% in extracting optical properties.

According to Heisenberg’s uncertainty principle, measurement of a quantum observable introduces noise to this observable and thus limits the available precision of measurement. Quantum non-demolition measurements are designed to circumvent this limitation and have been demonstrated in detecting the photon flux of classical light beam. Quantum non-demolition measurement of a single photon is the ultimate goal because it is of great interest in fundamental physics and also a powerful tool for applications in quantum information processing. This chapter presents a brief introduction of the history and a review of the progress in quantum non-demolition measurement of light. In particular, a detailed description is presented for two works toward cavity-free schemes of quantum non-demolition measurement of single photons. Afterward, an outlook of the future in this direction is given.

Time-correlated single-photon counting (TCSPC) applications usually deal with a high counting rate, which leads to a decrease in the system efficiency. This problem is further complicated due to the random nature of photon arrivals making it harder to avoid counting loss as the system is busy dealing with previous arrivals. In order to increase the rate of detected photons and improve the signal quality, many parallelized structures and imaging arrays have been reported, but this trend leads to an increased data bottleneck requiring complex readout circuitry and the use of very high output frequencies. In this paper, we present simple solutions that allow the improvement of signal-to-noise ratio (SNR) as well as the mitigation of counting loss through a parallelized TCSPC architecture and the use of an embedded memory block. These solutions are presented, and their impact is demonstrated by means of behavioral and mathematical modeling potentially allowing a maximum signal-to-noise ratio improvement of 20 dB and a system efficiency as high as 90% without the need for extremely high readout frequencies.

This paper presents a Fixed Priority Arbiter (FPA) which fully takes benefice of asynchronous design. It signs up in a Time Correlated Single Photon Counting system where the data transfer from the Time to Digital Converters to the readout or process unit is a critical aspect. In order to increase the photon counting rate of the setup, the device has to handle a large amount of data following the increasing number of sensors. Thanks to its asynchronous design, the proposed FPA can be easily assembled to build a wide data path tree to guide all the data to a single output. The implementation on a 180 nm CMOS technology shows that a 128 input wide path tree can achieve a burst input rate of 21 Giga event per second while sustaining an output data rate of 333 Mega event per second.

—This paper presents a new method to get real time hyperspectral images using time modulation of light and demodulation by means of General-Purpose computing on Graphics Processing Units (GPGPU). Three different Compute Unified Device Architecture (CUDA) implementations of real time hyperspectral images demodulation are presented. These methods are compared using a numerical simulation. The results show an execution time as low as 18 µs per wavelength and per frame for the Custom-made proposed implementation for a 512x512 pixels frame.

This paper presents the simulations and characterizations results of a hybrid Time to Digital Converter (TDC) fabricated in 180 nm standard CMOS. The design combines the traditional Analog Time to Amplitude Converter (TAC) and Digital TDC techniques to obtain a high adjustable time precision. These approach leads to a 3 bits enhancement of the least significant bit resolution (LSB) for the proposed design. The characterization results showed a time precision of 10 ps with an estimated INL of 5.6 ps rms for a 2.5 ns reference period clock and a 32 cells delay line loop.

We present the performance characteristics of a single photon avalanche diode (SPAD) fabricated in a 180 nm standard CMOS image sensor technology. The SPAD structure was implemented in 8 different diameters between 5 and 40 μm to determine the influence of size variation on the SPAD performances in terms of dark count rate, afterpulsing, efficiency and time resolution. The measurements show a dark count rate below 10 kHz at 15 °C with a low afterpulsing probability (0.2 % at an excess bias of 300 mV), a good Photodetection efficiency (~20 %) and a very good time resolution (<70 ps FWHM at 450 nm) .

— This paper introduces a fast and efficient method to characterize a Time to Digital Converter (TDC) transfer function using Poisson process events. We propose a correction method appropriate for ASIC as well as discrete or FPGA TDCs, to be implemented on a post process unit located after the Time to Digital Converter. We then apply the presented method to a case study that demonstrates the efficiency of the correction by measuring the fluorescence lifetime of a test fluorophore by time-correlated single photon counting (TCSPC) with and without correction. The results show a much nicer signal and better fitting with a 3-fold improvement of the fluorescence lifetime accuracy.